Enhancing the Performance of Heterogeneous Networks Using Optimized Cluster-Based Algorithm

Abstract

AbstractWith an ever-increasing number of user equipment (UEs) and rising bandwidth demands of new applications, deployment dense heterogeneous cellular networks have been embraced in various network scenarios. The cells experience unloaded due to the random UEs mobility and cells deployment, which degrades the network performance such as handover success, throughput, and load distribution. We propose an enhanced cluster-based algorithm for small cell mobility load-balancing networks to address such a problem. The conventional mobility load-balancing (MLB) algorithms consider only the contiguous neighboring cells and do not expand enough performance of the network, while other MLB algorithms consider the neighboring cells of the total network experienced unneeded MLB actions. The proposed load-balancing algorithm studies overloaded cells and neighbors using the proposed efficiency parameter. To begin with and to identify the overloaded cells, we propose an efficiency factor B that compares a pre-defined threshold and the network threshold to control the algorithm triggering by modifying the CIO parameters of the cluster cells in both medium-loaded and overloaded cells triggering in both medium-loaded and overloaded cells. Then, to control the distribution, we propose a method to shift only a portion of the serving cells load, so the target cell load after handover always be equal to or less than B. In a low UE speed scenario, the simulation results showed a lower standard deviation (SD) by 19.99% and enhanced throughput by 7.355%.KeywordsSelf-organizing networkSmall cellMobility load balancingHetNetsThroughputHandoverPhysical resource block